Instructions to use aglaia-models/surya-det-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use aglaia-models/surya-det-mlx with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir surya-det-mlx aglaia-models/surya-det-mlx
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
Surya Text-Line Detection — MLX
MLX port of surya's EfficientViTForSemanticSegmentation text-line detector
(EfficientViT-L backbone + SegFormer head). Box-F1 vs the original ~0.99-1.00 on
sample documents (see bench/RESULTS.md). Loads torch-free via this project's
surya2_mlx.detection.
Changes from the original
Weights converted to MLX (NHWC) format. No fine-tuning. Architecture and source: https://github.com/datalab-to/surya (Apache-2.0 code; weights under datalab terms).
- Downloads last month
- 19
Model size
38.4M params
Tensor type
F32
·
Hardware compatibility
Log In to add your hardware
Quantized
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support